ABSTRACT In high tuberculosis (TB) incidence settings, individuals with TB are often infected with multiple strains of M. tuberculosis complex (Mtbc). Despite this known fact, current TB transmission studies largely ignore within-host Mtbc heterogeneity. We believe that accounting for within-host Mtbc heterogeneity will reduce sampling bias and significantly improve the understanding of TB transmission dynamics in households and in community gathering places. For example, TB transmission studies in high TB incidence settings have found that the majority of TB cases occurring concurrently within the same household have non-matching molecular fingerprints. This finding has led to the conclusion that the majority of household TB cases were acquired outside of the household. However, none of those studies have appropriately accounted for within-host heterogeneity, which could have led to missed detection of Mtbc genetic clusters within the household. In addition, despite numerous TB transmission studies, factors that predict TB transmission remain poorly understood. Accounting for within-host Mtbc heterogeneity could improve the detection of pathogen and host related factors that affect transmission. Together, improved understanding of these areas could lead to more accurate identification of transmission networks and disease hotspots, which can then guide interventions to interrupt TB transmission. Moreover, the proposed research could instigate significant changes in the practice of future TB transmission studies by evaluating the impact of accounting for within-host Mtbc heterogeneity on TB transmission inference. The proposed research will address the current gaps in knowledge by incorporating two novel methods for detecting within-host Mtbc heterogeneity: 1) we will conduct whole-genome sequencing (WGS) on early primary culture samples to detect heterogeneous Mtbc strains; and 2) we will perform targeted amplicon-based sequencing of 150 genetic loci important for phylogenetic and resistance prediction. We will use advanced bioinformatic methods to integrate these sources of data on Mtbc heterogeneity. The proposed research will also utilize community-based door-to-door active case finding to minimize sampling bias. These methods will be applied to achieve 2 specific aims: 1) to determine the impact of accounting for within-host heterogeneity of Mtbc strains on inference in a population-based TB transmission study; and 2) to determine more accurately the proportion of household TB cases that are attributable to transmission within the household by conducting a prospective household contact study. We will also determine pathogen and host factors that predict individual and population-level transmission. This project will generate important scientific knowledge of TB transmission and factors that affect transmission, and findings will inform and guide targeted interventions to combat TB epidemics by interrupting the transmission network in local settings.